Object-Oriented Programming has well established design principles, such as SOLID. For many developers architecture and functional programming are at odds with each other: they don’t know how their existing tricks of the trade convert into functional design. This problem becomes worse as hybrid languages such as Scala, Java 8 and Ruby become common. I will talk about how functional programming helps you implement the SOLID principles, and how a functional mindset can actually help you achieve the holy grail of OO, encapsulation.
2. What on earth are you talking about?
SOLID Principles
Design Patterns
Anthropology
3.
4. In Quotes ...
"I find OOP technically unsound”
- Alexander Stepanov
OOP prevents any one programmer from "doing too much damage."
- Paul Graham
"OOP is to writing a program, what going through airport security is to flying"
- Richard Mansfield
"Brain explosion is like a traditional pasttime in #haskell"
"Some people claim everything is lisp. One time I was eating some spaghetti
and someone came by and said: 'Hey, nice lisp dialect you're hacking in there'"
"TDD replaces a type checker in Ruby in the same way that a strong drink
replaces sorrows."
- byorgey
7. What on earth are you talking about?
SOLID Principles
Design Patterns
Anthropology
8. SOLID Principles
● Basic Object Oriented Programming
Principles
● Make programs easier to maintain
● Guidelines to remove code smells
9. Single Responsibility Principle
● Each class/method should have single
responsibility
● Responsibility means “reason to change”
● The responsibility should be encapsulated
10. int countPrimes(int upTo) {
int tally = 0;
for (int i = 1; i < upTo; i++) {
boolean isPrime = true;
for (int j = 2; j < i; j++) {
if (i % j == 0) {
isPrime = false;
}
}
if (isPrime) {
tally++;
}
}
return tally;
}
11. int countPrimes(int upTo) {
int tally = 0;
for (int i = 1; i < upTo; i++) {
if (isPrime(i)) {
tally++;
}
}
return tally;
}
boolean isPrime(int number) {
for (int i = 2; i < number; i++) {
if (number % i == 0) {
return false;
}
}
return true;
}
14. Higher Order Functions
● Hard to write single responsibility code in
Java before 8
● Single responsibility requires ability to pass
around behaviour
● Not just functions, M&S higher order
functions
17. OCP as Polymorphism
● Graphing Metric Data
○ CpuUsage
○ ProcessDiskWrite
○ MachineIO
● GraphDisplay depends upon a
TimeSeries rather than each individually
● No need to change GraphDisplay to add
SwapTime
18. OCP by Functions as Arguments
// Example creation
ThreadLocal<DateFormat> formatter =
withInitial(() -> new SimpleDateFormat());
// Usage
DateFormat formatter = formatter.get();
// Or ...
AtomicInteger threadId = new AtomicInteger();
ThreadLocal<Integer> formatter =
withInitial(() -> threadId.getAndIncrement());
19. OCP as Immutability
● Immutable Object cannot be modified after
creation
● Safe to add additional behaviour
● New pure functions can’t break existing
functionality because it can’t change state
20. Liskov Substitution Principle
Let q(x) be a property provable about objects
x of type T. Then q(y) should be true for
objects y of type S where S is a subtype of T.
* Excuse the informality
21.
22. A subclass behaves like its parent.
* This is a conscious simplification
23. 1. Where the parent worked the child should.
2. Where the parent caused an effect then the
child should.
3. Where parent always stuck by something
then the child should.
4. Don’t change things your parent didn’t.
24. Functional Perspective
● Inheritance isn’t key to FP
● Lesson: don’t inherit implementation and
LSP isn’t an issue!
● Composite Reuse Principle already
commonly accepted OOP principle
26. Factory Example
interface Worker {
public void goHome();
public void work();
}
AssemblyLine requires instances of
Worker: AssemblyWorker and Manager
but a Robot doesn’t goHome()
27. Nominal Subtyping
● For Foo to extend Bar you need to see Foo
extends Bar in your code.
● Relationship explicit between types based
on the name of the type
● Common in Statically Typed, OO languages:
Java, C++
31. Structural Subtyping
● Relationship implicit between types based
on the shape/structure of the type
● If you call obj.getFoo() then obj needs a
getFoo method
● Common in wacky language: Ocaml, Go,
C++ Templates, Ruby (quack quack)
33. def addWorker(worker: {def work(step:ProductionStep)}) {
workers += worker
}
def newLine() = {
val line = new AssemblyLine
line.addWorker(new Manager())
line.addWorker(new StructuralWorker())
line.addWorker(new Robot())
line
}
34. Hypothetically …
def addWorker(worker) {
workers += worker
}
def newLine() = {
val line = new AssemblyLine
line.addWorker(new Manager())
line.addWorker(new StructuralWorker())
line.addWorker(new Robot())
line
}
35. Functional Interfaces
● An interface with a single abstract
method
● By definition the minimal interface!
● Used as the inferred types for lambda
expressions in Java 8
36. Thoughts on ISP
● Structural Subtyping obliviates the need for
Interface Segregation Principle
● Functional Interfaces provide a nominalstructural bridge
● ISP != implementing 500 interfaces
38. Dependency Inversion Principle
● Abstractions should not depend on details,
details should depend on abstractions
● Decouple glue code from business logic
● Inversion of Control/Dependency Injection is
an implementation of DIP
43. DIP Summary
● Higher Order Functions also provide
Inversion of Control
● Abstraction != interface
● Functional resource handling, eg withFile
in haskell
46. What on earth are you talking about?
SOLID Principles
Design Patterns
Anthropology
47. Command Pattern
• Receiver - performs the actual work.
• Command - encapsulates all the information
required to call the receiver.
• Invoker - controls the sequencing and
execution of one or more commands.
• Client - creates concrete command instances
51. public class Open implements Action {
private final Editor editor;
public Open(Editor editor) {
this.editor = editor;
}
public void perform() {
editor.open();
}
}
52. public class Macro {
private final List<Action> actions;
…
public void record(Action action) {
actions.add(action);
}
public void run() {
actions.forEach(Action::perform);
}
}
54. The Command Object is a Function
Macro macro = new Macro();
macro.record(editor::open);
macro.record(editor::save);
macro.record(editor::close);
macro.run();
68. Programming 80s style
● Strongly multiparadigm languages
○ Smalltalk 80 had lambda expressions
○ Common Lisp Object System
● Polyglot Programmers
● Fertile Language Research
● Implementation Progress - GC, JITs, etc.
70. 90s and 2000s Market Convergence
● Huge Java popularity ramp
○ Javaone in 2001 - 28,000 attendees
○ Servlets, J2EE then Spring
● Virtual death of Smalltalk, LISP then Perl
● Object Oriented Dominance
72. Increasingly Multiparadigm
● Established languages going multiparadigm
○ Java 8 - Generics + Lambdas
○ C++ - Templates, Lambdas
● Newer Languages are multi paradigm
○ F#
○ Ruby/Python/Groovy can be functional
○ New JVM languages:
■ Scala
■ Ceylon
■ Kotlin